An increasing number of distributed applications process continuously flowing data from distributed sources by applying queries to the data before distributing the data to geographically distributed recipients. An event stream processing engine (ESPE) continuously applies the queries to the data as it is received and determines which entities receive the processed data.